THE IMPACT OF CELEBRITY AND NON-CELEBRITY ENDORSER CREDIBILITY IN THE ADVERTISEMENT ON ATTITUDE TOWARDS ADVERTISEMENT, ATTITUDE TOWARDS BRAND, AND PURCHASE INTENTION.

ICEBESS 2016 Proceeding

THE IMPACT OF CELEBRITY AND NON-CELEBRITY ENDORSER
CREDIBILITY IN THE ADVERTISEMENT ON ATTITUDE
TOWARDS ADVERTISEMENT, ATTITUDE TOWARDS BRAND,
AND PURCHASE INTENTION
Steffi Tanjung1, Herry Hudrasyah1
1

Bandung Institute of Technology (School of Business and Management), Indonesia
Email: steffi.tanjung@sbm-itb.ac.id

Abstract
In this highly competitive era, consumers can be exposed to large volume of
advertisement. Hence, the concept of celebrity endorsement is commonly used to help
advertisement stands out from the surrounding media clutter in the communication
process. Numerous researchers have proved empirically the effectiveness and the
positive influence of the celebrity endorsement in advertising. However, the use of
celebrity endorsers can also inflict several potential disadvantages. As an alternative,
the use of non-celebrity endorsers as unknown individual should also be considered
since they will consequently intensify the focus on the product, give a chance for high

degree of control for company, cost advantage as well as possibility to have a better
fit between the product and endorsers. Therefore, this research is undertaken to
analyze the impact of celebrity and non-celebrity endorser credibility • attractiveness,
trustworthiness and expertise in the advertisement with respect to attitude toward
advertisement, attitude toward brand, and purchase intention of Indonesian
consumers, particularly city of Bandung and Jakarta. A questionnaire was distributed
to 384 women in Bandung and Jakarta; a pilot study was conducted for the
questionnaire improvement. Multiple regression analysis is applied to analyze the
relationship between the variables. This paper describes that celebrity endorser is
perceived to be more attractive while non-celebrity is perceived to be more trustworthy
and expert in significantly influencing consumer’s attitude toward advertisementbrand, and purchase intention. This result is in the consequence of several potential
disadvantages of using celebrity endorsement which have been mentioned above.
Keywords: Attitude toward Advertisement; Attitude toward Brand; Celebrity
Endorsement; Non-Celebrity Endorsement; Purchase Intention

INTRODUCTION
Consumers can be exposed to large volume of advertisements in this highly competitive era.
The number of advertisement which exceeds the acceptance level of a consumer in an
editorial media vehicle is considered as clutter and is frequently perceived as an undesirable
phenomenon both for advertisers and consumers (Ha L. , 1996). Consumers will pay less

attention to advertisements in a cluttered media environment because they will be irritated
by the advertisements and subsequently avoid the ads altogether. Other than that, the
excessive amount of advertisements presented will lead to the inability of consumers to
remember the ads due to their limited memory capacity (Ha & McCann, 2008). Therefore,
advertisers should pay attention to their advertising strategies that must be able to capture
consumer’s attention and influence their buying decision/ Delebrity endorsement such as
actors, actresses, entertainers and athletes as one of major advertising strategies is commonly
International Conference on Ethics of Business, Economics, and Social Science

|| 231

ISSN: 2528-617X

used for promotional activities. A celebrity is a person who has a salient profile and owns a
substantial degree of public fascination and influence in daily media (Asad, Hayat, &
Mehmood, 2013). It is affirmed by (Muda, Musa, & Putit, 2012) that celebrity can help
advertisement stand out from the surrounding media clutter by cutting off excess clutter in
the communication process thus the communicative capability is improved.
However, there are also several potential risks involved of using celebrity
endorsement/ Delebrity endorsement can inflict high risk and ‘no gain’ situations such as the

‘scandals’ surroundings celebrities like Nichael Kackson- Late Noss- Critney Tpears- Qaris
Hilton due to the limited control over the celebrity’ persona by the company (Roozen &
Claeys, 2010) and also celebrity’s future behavior (Till & Shimp, 1998). It is also stated by
Till & Shimp (1998) that any negative information which is produced by the celebrity’s
private actions will negatively affect the company, perception of consumers about the
celebrity and also the endorsed brand. Moreover, empirical studies have found that the
existence of celebrity can attract consumers to watch the advertisement but consumers will
fail to notice the brand which being promoted because they focus on the celebrity (Erdogan,
1999).
Consequently, the trend of using non-celebrity advertising is growing due to the
negative effect of celebrity endorsement which could damage the brand image (Saeed,
Naseer, Haider, & Naz, 2014). According to (Prieler, Kohlbacher, Hagiwara, & Arima,
2010), non-celebrities persons are used in the advertisement in order to keep away the
negative effect of celebrity endorsement. Non-celebrity is an individual who isn’t wellknown, chosen by company based on the demographics of existing target market (Rodriguez,
2008). As an alternative, companies can create their own endorsers i.e. a created
spokesperson using not so well-known individuals in order to give a chance for high degree
of control (Roozen & Claeys, 2010) cost advantage and also possibility to have a better fit
between the product and endorsers (Erdogan, 1999); (Tom, et al., 1992).
Numerous researches have proved empirically the effectiveness and the positive
influence of the endorsement by the celebrities in advertising- specifically on endorser’s

credibility, message reminder, announcements approval and purchase intention (Menon,
2001); (Pornpitakpan, 2003); (Pringle & Binet, 2005); (Roy, 2006). Besides the fact of their
potential advantages, using celebrity endorsement can cause several potential risks as
mentioned above. All of those arguments lead us to notice that the use of non-celebrity can
be also another option which should be considered seriously. Choosing the right endorser is
both a crucial and risky decision, thus it is needless to argue that effectiveness of both
celebrity and non-celebrity endorsement is well worth to be further analyzed.
However, the previous studies regarding the impact of celebrity and non-celebrity
endorsement have been conducted in different countries and consumer groups. The studies
in Indonesia about that topic are still limited, thus this research may generate different result
from the earlier study which can be considered for advertisers for further improvement. The
credibility of endorsers were considered as an important factor which might impact
consumer’s purchase intention and attitudes toward advertising Therefore, this research aims
232 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

to find out the impact of credibility dimensions of celebrity and non-celebrity endorser •
attractiveness, trustworthiness, and expertise on attitude towards advertisement, attitude
towards brands, and also purchase intention of Indonesian consumers, particularly city of

Bandung and Jakarta. Moreover, the result of this research can reveal which attributes of
celebrity and non-celebrity that most influence the attitude toward advertisement, brands,
and also purchase intention that will provide strategic advantage about selecting the right
endorsers.
LITERATURE REVIEW
Endorsements as Part of Advertising Strategy
Endorsements are used to grab attention to the product since the endorsers can transfer its
characteristics onto the product and if the target consumers like or aspire to have the
endorser’s characteristics- they would like to purchase the product of service (Byrne,
Whitehead, & Breen, 2003). When choosing an endorser, advertisers can either choose a
celebrity or created spokesperson (Tom, et al., 1992). According to (van der Waldt, van
Loggerenberg, & Wehmeyer, 2009), there are two kinds of created spokesperson; either real
(unknown) people acting out or animated / imaginary roles. In this study, the real people is
considered as the created endorsers and called as non-celebrity endorsers.
Celebrity Endorsement
Bccording to NcDracken )2;9;*- •B celebrity endorser is a person who enjoys the public
recognition and who uses this recognition in the name of goods while appearing with this
one in advertising•/ Delebrities are frequently used by companies because they can easily
escalate the brands of the company as well as save resources in building credibility by
transferring their values to the brand (Byrne, Whitehead, & Breen, 2003). Other than that,

due to the attractive and pleasant qualities of celebrity, consumers will associate the brand
with their favorite celebrity that make the advertising more effective (Atkin & Block, 1983).
It is stated by (Dyer, 1986), that celebrity is one of the most effective tools to gain consumer’s
attention and get him or her to interpret the appropriate message in a limited amount of time
and space.
Non-Celebrity Endorsement
•B non-celebrity is a person who, prior to placement in the campaign, has no public notoriety
but appears in an advertisement for the product• (Menon, 2001). Non-celebrity endorsers as
unknown person will consequently intensify the focus on the product or brand endorsed
rather than on the endorsers themselves because they are usually more representative for the
targeted market compared to the celebrity endorsers (Brett, Wentzel, & Tomczak, 2008).
The non-celebrity credibility became the main asset in order to increase advertising
effectiveness and also leads more positive attitudes than celebrity (Gaied & Rached, 2010)

International Conference on Ethics of Business, Economics, and Social Science

|| 233

ISSN: 2528-617X

The Source Credibility Model
In the literature, two general models are often used to analyze celebrity endorsement: the
source credibility model and the source attractiveness model. The Source Credibility Model
was found by Hovland et. al (1953) while The Source Attractiveness Model was proposed
by NcHuire )2;96* which mainly based on four dimensions- namely •familiarity••likability•- •similarity•- and •attractiveness•/ B source is a person or an organization
disseminating a message (Gaied & Rached, 2010). According to Hovland et al (1953), the
Source Credibility Model fundamentally states that the effectiveness of a message depends
on perceived level of expertise and trustworthiness of an endorser or the source. Therefore,
two fundamental dimensions of source credibility are expertise and trustworthiness. Besides
those two dimensions, it is stated that attractiveness of the source is also accepted as a
dimension of credibility (Ohanian, 1990). The source familiarity, likability and similarity
were not used in this research. Hence, there are three dimensions of source credibility:
‘Bttractiveness’- ‘Urustworthiness’ and ‘Fxpertise’ which described as follow<
Attractiveness
Attractiveness is the stereotype of positive associations to a person and not only involves
physical attractiveness but also the other characteristics such as personality and athletic
ability (Erdogan, 1999). When consumers feel that there is something in common between
them and the endorser, they will find the endorsers as attractive. Attractive endorsers are
more successful in changing beliefs (Ohanian, 1990). Perceived attractive endorsers are
more likely to lead to the purchase intention (van der Waldt, van Loggerenberg, &

Wehmeyer, 2009). The attractive characters in the advertisement induce a more favorable
attitude towards the advertising and a high purchasing intentions (Petroshius & Crocker,
1989).
Trustworthiness
Urustworthiness is •the honesty- integrity- and believability of an endorser• )Frdogan et al2001). Trustworthiness refers to level of confidence that consumers have regarding the
validity of assertions communicated by the endorser (Ohanian, 1990). It becomes important
to choose the endorsers with high level of trustworthiness, honesty, and affinity which would
further enhance the advertisement (Shimp, 2003). Celebrity are perceived to be more expert
and trustworthy than non-celebrity. However, it was stated that attractiveness and
trustworthiness of celebrity are not related to the purchase (Ohanian, 1990).On the other
side, Pornpitakpan (2003) stated that all of three credibility dimensions were positively
related to the purchase intention.
Expertise
Expertise refers to the degree to which the endorser is perceived to have the adequate
knowledge, experience or skills to promote the product (van der Waldt, van Loggerenberg,
& Wehmeyer, 2009)/ Jt’s not important whether the endorser own an expertise or not as long
234 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

as consumers perceived it to be so (Erdogan, 1999). According to Ohanian (1990), expertise
was significantly related to purchase and the best assessment of the endorser’s expertise level
is the knowledge about product. If the endorser has more knowledge about the product, it
will lead to positive attitude toward the brand (Chan et al, 2013). Celebrities are perceived
to have more expertise than the created spokesperson (van der Waldt, van Loggerenberg, &
Wehmeyer, 2009).
Attitude towards Advertisement (AAD)
It was found that endorser credibility has an impact on the Attitude towards Advertisement
(AAD) (Lafferty & Newel, 2000). Therefore, the AAD would be one of the dependent
variables in this study. Attitude towards Advertisement (AAD) refers to the behavior of the
audience towards the advertising (Bauer & Greyser, 1968) which can be represented through
consumer’s favorable or unfavorable reaction towards a particular advertisement
(Mackenzie and Lutz, 1989). The cognitive ability towards the advertising of consumer
would be reflected in their thoughts and feelings which subsequently will influence their
attitude towards advertising/ Ience- the consumers’ attitude toward advertising became one
of the influential indicator for measuring advertising effectiveness (Mehta, 2000);
(Mackenzie and Lutz, 1989).
Attitude towards Brand (AAB)
Bttitude towards brand )BBC* is defined as an individual’s internal evaluation of the
brand/• (Mitchell & Olson, 1981). Attitude toward brand is a predisposition to responds in a

favorable or unfavorable way to a particular brand after the individual has been shown the
advertising stimulus (Phelps & Hoy, 1996). The earlier studies stated that AAD and AAB
are not considered as one variable since they are different from each other and had been used
as separated variables (e.g. (Goldsmith & Newell, 2000)).
Purchase Intention
Purchase intention would be one of dependent variables since it is claimed by )P‘Nahony
& Meenaghan, 1998) that endorsers’ credibility had the greatest impact on the purchase
intention. Purchase intention is personal behavioral tendency to a particular product or brand
(Bagozzi & Burnkrant, 1979) and the probability a customer will purchase a particular
product (Fishbein & Ajzen, 1975); (Dodds, Monroe, & Grewal, 1991) (Schiffman & Kanul,
2000). Intentions are different with attitudes. TV commercials, newspaper or magazine
advertisings can be used to push the exposure rate of a product and the brand attitude and
knowledge of consumers in order to increase the purchase intention (Miciak & Shanklin,
1994).

International Conference on Ethics of Business, Economics, and Social Science

|| 235

ISSN: 2528-617X

Theoretical Framework and Research Hypothesis

Figure 1 Theoretical Framework
Based on the literature review and the research objectives, the research hypotheses are
constructed as follows:
H1: Celebrity Attractiveness has a significant impact on:
H1.1 the attitude toward advertising
H1.2 the attitude toward brand
H1.3 purchase intention
H2: Celebrity Trustworthiness has a significant impact on:
H2.1 the attitude toward advertising
H2.2 the attitude toward brand
H2.3 purchase intention
H3: Celebrity Expertise has a significant impact on:
H3.1 the attitude toward advertising
H3.2 the attitude toward brand
H3.3 purchase intention
H4: Non-Celebrity Attractiveness has a significant impact on:
H4.1 the attitude toward advertising
H4.2 the attitude toward brand
H4.3 purchase intention
236 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

H5: Non-Celebrity Trustworthiness has a significant impact on:
H5.1 the attitude toward advertising
H5.2 the attitude toward brand
H5.3 purchase intention
H6: Non-Celebrity Expertise has a significant impact on:
H6.1 the attitude toward advertising
H6.2 the attitude toward brand
H6.3 purchase intention
METHODOLOGY
Sampling
This research applied non-probability sampling which is purposive sampling method.
According to purposive sampling method, the criteria of respondents should be women aged
16-30 years old who live in Bandung or Jakarta. In determining the sample size for this
research, the researcher refers to Sekaran (2003) that stated if the population of the study
exceeds a million with the level of confidence 95% and margin error of 5%, sample of 384
respondents would be enough to represent the population. Hence, 384 women were targeted
as the respondent for this research.
Data collection
Data is collected from both primary and secondary sources. Primary data was collected using
quantitative method by distributing self-administered online questionnaire to the targeted
respondents during July 2016. Pilot study was conducted to 30 people for the evaluation of
questionnaire and also further improvement before the questionnaire was fully distributed.
Meanwhile, secondary data was obtained through journals, textbooks, and also online
resources which can be used to develop the hypothesis.
Questionnaire Design
Before filling out the survey, respondents were exposed to two pieces of advertisements. The
videos of both advertisements were featured in the second section of questionnaire and the
respondents were asked to watch the videos since the entire questions would be refer to those
videos. In addition, some scenes in both videos were also presented in form of picture in
every section of questionnaire so that the respondents would able to keep those
advertisements in their mind while filling out the questions/ Qantene’ shampoo
advertisement which employed celebrity endorser was selected. The other selected
advertisement was a testimonial advertising of ‘Eove’ shampoo which featured three
‘regular’ women as non-celebrity endorsers. Both type of endorsers played a different role
in conveying the messages in those advertisement although both of them were jointly

International Conference on Ethics of Business, Economics, and Social Science

|| 237

ISSN: 2528-617X

explaining the benefits and their experiences of using the product/ Saline Thah in ‘Qantene’
advertisement act as celebrity endorser as well as brand ambassador of Pantene while the
‘regular’ women in ‘Eove’ advertisement act as typical ordinary user/
In terms of advertised products, the researcher decided to opt similar type of product
which is hair fall shampoo. The researcher wished to eliminate any possible effect of brand
familiarity and brand loyalty that might have been existed and varied in the mind of
respondents. Hence, the researcher decided to choose shampoo product category which
included in Fast Moving Consumer Goods (FMCG) that generally requires low level of
involvement. According to Petty et al (1983), a person will not expend the effort or engage
in any extensive thought to think about the product-relevant arguments in the advertisement,
but may instead focus on the attractiveness- credibility- or prestige of the product’s endorser
(peripheral route) when considering low involvement product.
The factor such as brand familiarity, brand loyalty and price of both advertisement
were assumed to be equal. In the beginning of the questionnaire, it has been also emphasized
that for the importance of the research the respondent was asked to put herself as an
individual who has never watched both presented advertisements and never recognized about
both brands in the advertisements. The price of both products in the advertisements are
assumed to be equal.
Measurement Instrument
The questionnaire of this study used Multiple Choice scale for the first section which
measured the demographic variable as much as five items while the rest of the sections used
a seven-point Mikert scale ranging from 2< •Ttrongly Eisagree• to 8< •Ttrongly Bgree•/ Jn
this study, in order measure endorser’s credibility- Phanian’s )2;;1* scale was used for both
celebrity and non-celebrity endorsers since Ohanian (1990) stated that the scale can be
applied beyond the use of celebrity spokespersons. It can be applied effectively in choosing
the most appropriate ‘average consumer’ as a spokesperson/ Uherefore- the dimensions of
source credibility were measured using six sub-variables, namely celebrity attractiveness,
non-celebrity attractiveness, celebrity trustworthiness, non-celebrity trustworthiness,
celebrity expertise, and non-celebrity expertise as the independent variables. Attitude
towards Advertisement (AAD) would be measured using four sub-variables adapted from
the previous research of Ling, Piew & Chai (2010)while Attitude towards Brand (AAB) was
measured using the scale of Mackenzie & Lutz (1989) which consists of three pairs of
descriptive words. The last dependent variable, namely Purchase Intention (PI) would be
evaluated using the scale of Batra & Ray (1986).
RESULTS
Demographic Profile of Respondents
The table 1 indicates that the respondents aged of 21-25 years old dominated with the
percentage of 71.9%. Other than that, majority of respondents were found to reside in
Bandung with the percentage of 82.6%. In terms of education background, it can be seen
238 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

that Diploma Degree with the percentage of 70.3% has the highest frequency among the
other education level. These results are in consequence of the survey distribution which
mainly distributed among high school and college students in Bandung in order to support
young population who have been claimed to be heavily influenced by celebrity endorsers
(Bush, Martin and Bush, 2004; Stanton, Currie, Oei, and Silva, 1996; Atkins and Block,
1983).
Table 1. Demographic Profile of Respondents

Age

City of Residence

Education
Background

Classification
16-20
21-25

Frequency (n)
99
276

Percentage (%)
25.8
71.9

26-30

9

2.3

Bandung

317

82.6

Jakarta

67

17.4

Middle School

56

14.6

High School

43

11.2

Diploma Degree
Bachelor Degree

270
8

70.3
2.1

Master Degree

7

1.8

Validity and Reliability Test
All indicators for each variable were tested for its validity and reliability. Researcher used
Kaiser-Meyer-Olkin (KMO) score and factor loading to test its validity of each indicator.
The factor loading of each item of indicator must be above 0.5 and KMO score of each
variable should be Š 1/6 to be regarded as valid. Meanwhile, in order to test the reliability of
each variable- the Dronbach’s Blpha test is used where the result should be more than 0.6 to
be regarded as reliable/ Uhe validity and reliability test of endorser’s credibility showed that
all of the constructs within that variable were qualified to be included for the regression test
since the factor loading of each construct is higher than 0.5 and the score of KMO reached
1-;15/ Gor the reliability test- Fndorser’s Dredibility variable was considered as reliable as
the Dronbach’s Blpha score was 1/;43/
The variable of Attitude toward Advertisement also managed to result KMO score at
0.818 and its entire factor loading was also found above 0.5. These results indicate that each
construct within that variable supported the variable as on component and can be further
examined. In terms of its reliability, Attitude toward Advertisement passed the reliability
test since the Dronbach’s Blpha was 1/;22/ Pther than that- Bttitude toward Crand and
Purchase Intention were also considered as valid since their KMO score are 0.754 and 0.5,
respectively and their entire factor loading exceeded 1/6/ Uhe Dronbach’s Blpha score of
both variables exceeded 0.6 with the score for Attitude toward Brand was 0.870 and the
International Conference on Ethics of Business, Economics, and Social Science

|| 239

ISSN: 2528-617X

score for Purchase Intention was 0.895. To sum up, it can be concluded that the entire
variables passed the validity and reliability test and can be further analyzed.
Relationship between Independent Variables and Dependent Variables
The hypothesis which were presented in the theoretical framework would be tested using
multilinear regression. Regression is used to analyze the influence factors have to other
variables. According to the framework, there are six regression models which will be
analyzed. The hypothesis testing would be conducted on each model through classical
assumption test and then followed by multiple linear regression. Classical assumption test
was performed on each regression model in order to determine the validity of the regression.
The classical assumption test consist of several tests, namely normality test using
Kolmogorov-Smirnov Lilliefors correction; multicollinearity test using Variance Inflation
Gactor )WJG* and tolerance value and heteroscedasticity test using Tpearman’s rank
correlation. According to the result, it can be concluded that the data of entire models have
passed all the classical assumption test and can be further analyzed using multiple linear
regression.

Table 2 Relationship between Independent Variables to Dependent Variable of Model 1
and Model 2
Mo
del

Dependent
Variable

1st

Attitude
towards
Advertiseme
nt (AAD)

2nd

Attitude
towards
Advertiseme
nt (AAD)

Independent
Variable
Celebrity
Attractiveness
Celebrity
Trustworthiness
Celebrity
Expertise
Non-Celebrity
Attractiveness
Non-Celebrity
Trustworthiness
Non-Celebrity
Expertise

ANOVA
F

123.137

113.860

Sig.

0.000

0.000

Hypothesi
s

Coefficient
R2

Beta

Sig.

0.404

0.002**

0.986

0.000***

0.489

0.000***

H3.1

-0.070

0.529

H4.1

1.145

0.000***

0.640

0.000***

H1.1
0.493

0.473

H2.1

H5.1
H6.1

Table 2 indicates that based on the significance level table, it can be seen that the
entire independent variables of both models strongly influence (***) the dependent variable
although Celebrity Attractiveness was found to have lower significance level (*) compared
to the other variables. However, Non-Celebrity Attractiveness showed its significance level
of 0.529 which indicates that Non-Celebrity Attractiveness has no significant influence on
Attitude towards Advertisement. The negative beta coefficient of Non-Celebrity
Attractiveness indicates that every increase of Non-Celebrity Attractiveness value and the
other two variables are remained constant will decrease the Attitude toward Advertisement
as much as 0.070 but not significantly, since its significance level was higher than 0.05.
240 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

Other than that, Non-Celebrity Trustworthiness has the biggest influence on Attitude toward
Advertisement )¥?2/256- p?1/111* and followed by Delebrity Urustworthiness )¥ ?1/;97p=0.000). Those results above are contrary with the previous research of Bhatt, Jayswal, and
Patel (2013) concluded that trustworthiness and expertise of celebrity endorser have no
significant impact on Attitude toward Advertisement while attractiveness was found to
significantly influence.
Table 3 Relationship between Independent Variables to Dependent Variable of Model 3
and Model 4
Mo
del

Dependent
Variable

3rd

Attitude
towards
Brand
(AAB)

4th

Attitude
towards
Brand
(AAB)

Independent
Variable
Celebrity
Attractiveness
Celebrity
Trustworthiness
Celebrity
Expertise
Non-Celebrity
Attractiveness
Non-Celebrity
Trustworthiness
Non-Celebrity
Expertise

ANOVA
F

75.682

72.788

Sig.

0.000

0.000

Hypothesi
s

Coefficient
R2

Beta

Sig.

0.386

0.000***

0.278

0.000***

0.212

0.001**

H3.2

-0.016

0.790

H4.2

0.516

0.000***

0.218

0.001**

H1.2
0.374

0.365

H2.2

H5.2
H6.2

Table 3 presents the result of multiple linear regression performed on the Model 3
and Model 4. All dimensions of both endorsers credibility in Model 3 and Model 4 have a
strong impact (***) on Attitude toward Brand except for the Non-Celebrity Attractiveness
which is found to have significance level above 0.05. Hence, it can be stated that NonCelebrity Attractiveness has no significant impact on Attitude toward Brand which rejected
the hypothesis H4.2. Meanwhile, Celebrity Expertise and also Non-Celebrity Expertise are
perceived to have a lower significant impact (**) on Attitude toward Brand compared to the
other variables in both models. Looking at the beta scores table, it can be seen that NonCelebrity Trustworthiness is the most influential variable on Attitude toward Brand
(¥=0.516, p=0.000) and followed by Celebrity Attractiveness (¥=0.386, p=0.000). This
finding is conflicting with the previous research of Bhatt, Jayswal, and Patel (2013) celebrity
endorser respectively while attractiveness was not playing any significant role. It can be
summarized that H1.2, H2.2, H3.2, H5.2 and H6.2 are confirmed.
Based on table 4, it can be seen that only Celebrity Trustworthiness and NonCelebrity Trustworthiness which were found to have strong influence (***) towards
Purchase Intention while Celebrity Attractiveness has a lower significant value (p=
0.017(*)). Sertoglu, Catli, Korkmaz (2014) argued that perceived attractiveness,
trustworthiness, expertise of both celebrity endorser and created spokesperson significantly
influence the purchase intention. However, the finding obtained from Model 5 and Model 6
presented a different result. It was found that Celebrity Expertise, Non-Celebrity
International Conference on Ethics of Business, Economics, and Social Science

|| 241

ISSN: 2528-617X

Attractiveness, and Non-Celebrity Expertise generated a significant value above 0.05 which
indicates that these three independent variables have no significant impact on Purchase
Intention. Instead, these results are in line with the findings of Gupta, Kishore, Verma (2015)
which claimed that attractiveness and trustworthiness of a celebrity endorser have a
significant impact on consumer purchase intention while expertise of celebrity endorser does
not impact significantly. In terms of beta coefficient, it can be concluded that Non-Celebrity
Trustworthiness is the most influential variable toward Purchase Intention ((¥=0.177,
p=0.000) followed by Celebrity Trustworthiness (¥=0.176, p=0.000). This also proves
Hupta)311*’ s finding that trustworthiness of celebrity endorser should be the most crucial
factor in making a choice of celebrity endorsers followed by attractiveness and expertise.
Summing up the analysis above, the hypothesis H1.3, H2.3 and H5.3 are supported whereas
hypothesis H3.3, H4.3, and H6.3 are rejected.
Table 4 Relationship between Independent Variables to Dependent Variable of Model 4
and Model 6
Mo
del

5th

6th

Dependent
Variable

Purchase
Intention

Purchase
Intention

ANOVA

Independent
Variable
Celebrity
Attractiveness
Celebrity
Trustworthiness
Celebrity
Expertise
Non-Celebrity
Attractiveness
Non-Celebrity
Trustworthiness
Non-Celebrity
Expertise

F

48.556

34.430

Sig.

0.000

0.000

Hypothesi
s

Coefficient
R2

Beta

Sig.

0.077

0.017*

0.176

0.000***

0.026

0.375

H3.3

-0.009

0.760

H4.3

0.177

0.000***

0.051

0.103

H1.3
0.277

0.214

H2.3

H5.3
H6.3

DISCUSSION AND CONCLUSION
Based on the result above, it can concluded that almost all dimensions of celebrity and noncelebrity endorser credibility have a positive significant impact on the Attitude toward
Advertisement, Attitude toward Brand and Purchase Intention. It was found that NonCelebrity Attractiveness does not significantly influence (p>0.05) the entire dependent
variables. In terms of beta coefficient, the entire beta coefficient of Non-Celebrity
Attractiveness were found to be negative. It indicates that the relationship between NonCelebrity Attractiveness and all dependent variables are inversely related which means every
increase of attractiveness of non-celebrity will reduce the impact towards the dependent
variables. Therefore, this study resulted a very important point to be notice that it is
unnecessary for advertisers to consider the attractiveness of non-celebrity in the endorser
since it will not significantly impact the advertising effectiveness. Moreover, endorser who
perceived to be more attractive might lead consumers fail to notice the brand being promoted
because they tend to be more focus on the endorser.
242 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

The results also shows that Celebrity Expertise and Non-Celebrity Expertise do not
have any significant impact towards Purchase Intention which are contrary with the findings
from previous research of Sertoglu, Catli, Korkmaz (2014) stated that perceived
attractiveness, trustworthiness, expertise of created spokesperson can influence the intention
to purchase in a significantly positive way and findings from Ohanian (1991) which
concluded that expertise of celebrity spokesperson was found to have significant impact on
purchase intention. Moreover, based on the descriptive analysis on this study, it can be stated
that larger group of respondents believed that non-celebrity endorsers were perceived to be
more trustworthy in conveying statement in ads compared to celebrity endorser. This is in
line with the literature regarding the potential overexposure effect of celebrity endorser. It
was claimed by Solomon et al (2002) that overexposure may induce a general perception of
consumers that celebrities endorse many brands only for the motive of money. Hence, the
result that Non-Celebrity Trustworthiness is found to have biggest influence on Attitude
toward Advertisement, Attitude toward Brand, and Purchase Intention are plausible. This
can be another major finding which can be considered by advertisers. Meanwhile in terms
of type of endorser, celebrity endorser is perceived to be trustworthy in generating Attitude
toward Advertisement and Purchase Intention while the attractiveness of celebrity endorser
has a more significant impact on Attitude toward Brand.
Limitations of the Study and Directions for Future Research
There are some certain limitations in this study which may provide direction for future
research. Some of the key areas for future research are describes as follow:
1. The data obtained was limited to women aged 16-30 years old who live in Bandung and
Jakarta, Indonesia. A study may be conducted in another city or country with another
group age in order to gain more comprehensive view of consumer perception.
2. This research is limited to certain advertisement and product. Future research may
investigate the impact of celebrity and non-celebrity endorser credibility with varying
level of advertisement and product.
3. Future research may develop this study by evaluating the relationship between Attitude
toward Advertisement, Attitude toward Brand, and also Purchase Intention which are not
discussed in this study.
REFERENCES
Asad, A., Hayat, M. N., & Mehmood, S. (2013). Comparative Analysis of Celebrity and
Non-Celebrity Advertisement. A Research Journal of Commerce, Economics, and
Social Sciences, 9.
Atkin, C., & Block, M. (1983). Effectiveness of celebrity endorsers. Journal of Advertising
Research , 23(1), 57-61.
Byrne, A., Whitehead, M., & Breen, S. (2003). The Naked Truth of Celebrity Endorsement.
British Food Journal 105 (4/5), 288-296.

International Conference on Ethics of Business, Economics, and Social Science

|| 243

ISSN: 2528-617X

Erdogan, B. (1999). Celebrity Endorsement: A literature review. Journal of Marketing
Management Vol.41 No.3, 291-314.
Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An Introdution to
Theory and Research. MA: Addison-Wesley.
Fournier, S. (1998). Consumers and Their Brands: Developing Relationship Theory in
Consumer Research. Journal of Consumer Research , 343-373.
Gaied, A. M., & Rached, K. S. (2010). The Persuasive Effectiveness of Famous and NonFamous Endorsers in Advertising. IBIMA Publishing.
Goldsmith, R. E., & Newell, S. J. (2000). The Impact of Corporate Credibility and Celebrity
Credibility on Consumer Reaction to Advertisements and Brands. Journal of
Advertising Vol. 29 No. 3, 43.
Ha, L. (1996). Observations. Advertising clutter in consumer magazines: dimension and
effects. Journal of Advertising Research, 36(4), 76-84.
Ha, L., & McCann, K. (2008). An Integrated Model of Advertising Clutter in Offline and
Online Media. International Journal of Advertising, 27(4), 570-571.
Hovland, C., Janis, I., & Kelly, H. (1953). Communication and Persuasion. New Haven,CI:
Yale University Press.
McCracken, G. (1989). Who is the celebrity endorser?Cultural foundations of the
endorsement process. Journal of Consumer Research 16(3), 310-322.
McGuire, W. (1985). Attitudes and Attitude Change, In: Handbook of Social Psychology.
NY: Random House.
Mehta, A. (1994). How Advertising Response Modelling (ARM) Can Increase Ad
Effectiveness. Journal of Advertising Research, 34(3), 62-74.
Menon, K. (2001). Celebrity advertising: an assessment of its relative effectiveness.
Miciak, A., & Shanklin, W. (1994). Choosing Celebrity Endorsers. Marketing Management
3(3), 50-59.
Mitchell, A. A., & Olson, J. C. (1981). Are Product Beliefs and the Only Mediator of
Advertising Effect on Brand Attitude? Journal of Marketing Research 18, 318-332.
Muda, M., Musa, R., & Putit, L. (2012). Breaking through the clutter in media environment:
how do celebrities help? Procedia-Social and Behavioral Sciences 42, 374-382.
P‘Nahony- T/- ' Neenaghan- U/ )2;;9*/ Uhe Jmpact of Delebrity Fndorsement on
Consumer. Irish Marketing 10(2), 112-113.
Ohanian, R. (1990). Construction and Validation of A Scale to Measure Celebrity Endorser's
Perceived Expertise, Trustworthiness, and Attractiveness. Journal of Advertising
19(3), 39-52.
Petroshius, S., & Crocker, K. (1989). An Empirical Analysis of Spokesperson
Characteristics on Advertisement and Product Evaluations. The Academy of
Marketing Science 17(3), 217-225.
Phelps, J. E., & Hoy, M. G. (1996). The Aad-Ab-PI Relationship in children: the impact of
brand familiarity and measurement timing. Psychology & Marketing, Vol. 13(1), 77101.
244 || International Conference on Ethics of Business, Economics, and Social Science

ICEBESS 2016 Proceeding

Pongiannan, K., & Chinnasamy, J. (2014). Do Advertisements for Fast Moving Consumer
Goods Create Response among the Consumers? International Journal of Innovation,
Management and Technology, Vol. 5, No. 4.
Pornpitakpan, C. (2003). The effect of celebrity endorsers perceived credibility on product
purchase intention: the case of singaporeans. Journal of International Consumer
Marketing Vol.16 No.2, 55-74.
Prieler, M., Kohlbacher, F., Hagiwara, S., & Arima, A. (2010). Older Celebrity Versus Non
• Celebrity Television Advertising: A Japanese Perspective. Keiv Communication
Review, 32.
Pringle, H., & Binet, L. (2005). How marketers can use celebrities to sell more effectively.
Journal of Consumer Behaviour March 4 (3), 201-214.
Rodriguez, K. (2008). Apparel brand endorsers and their effects on purchase intentions: a
study of Phhillipine consumers. Phillipine Management Review 15, 83-99.
Roozen, I., & Claeys, C. (2010). The relative effectiveness of celebrity endorsement for print
advertisement. Review of Business and Economics, 77.
Roy, S. (2006). An exploratory study in celebrity endorsments. Journal of Creative
Communications Vol.1 No.2. New Delhi: Sage Publications.
Saeed, R., Naseer, R., Haider, S., & Naz, U. (2014). Impact of celebrity and non-celebrity
advertisement on consumer perception. The business & management review, Vol.4
No.3, 156.
Till, B., & Shimp, T. (1998). Endorsers in Advertising: The Case of Negative Celebrity
Information. Journal of Advertising, 27(1), 67-82.
Tom, G., Clark, R., Elmer, L., Grech, E., Masetti, J., & Sandhar, H. (1992). The use of
created versus celebrity spokespersons in advertisement. The journal of consumers
marketing Vol.9 No.4, 45-51.
van der Waldt, D., van Loggerenberg, M., & Wehmeyer, L. (2009). Celebrity endorsement
versus created spokespersons in advertising: a survey among students. SAJEMS NS
12 No 1, 103.

International Conference on Ethics of Business, Economics, and Social Science

|| 245